Timing a Telegram Drip Sequence: Why Email Intervals Are Wrong Here
Timing a Telegram Drip Sequence: Why Email Intervals Are Wrong Here
Most Telegram sequences are built by someone copying an email sequence: day 1, day 3, day 7, day 14.
Those intervals encode an assumption that does not hold. Email intervals are spaced in days because inboxes are checked in batches — a message sent Tuesday might be seen Thursday, so you leave room. The gap absorbs uncertainty about when the message will be seen.
Telegram is a push notification on a device in someone's hand. There is no inbox batching. Some uncertainty remains — people mute chats, archive them, leave notifications off, and you get no read receipts to tell you which — but it is far smaller than email's, and it is not what the gap is there to absorb. The interval is doing a completely different job, and copying the email number gets it wrong in both directions.
What the interval is actually for
On Telegram, the gap between messages is managing two things:
Attention. Too fast feels like spam. Too slow and they have forgotten who you are — which happens faster on a messaging app, because your message sits among conversations with actual humans.
Blocks. This is the real constraint. Telegram has no unsubscribe link. Irritation converts directly into a block, and a block is permanent from your side — you cannot re-open the conversation, only the user can. Telegram does tell you when it happens: my_chat_member fires in private chats precisely when the bot is blocked or unblocked, and it arrives without any allowed_updates opt-in. Since you have no way to ask for an unblock, budget as if it were permanent. On email, over-sending costs you an unsubscribe. Here it costs you the person.
So the interval is not a delivery consideration. It is a tolerance budget, and the currency is durable reach.
A shape that works
Front-loaded, then decaying:
| Message | Timing | Job |
|---|---|---|
| 1 | Immediately | Deliver what they came for |
| 2 | +1 to 4 hours | One useful thing, no ask |
| 3 | +1 day | The problem you solve |
| 4 | +2 days | The offer |
| 5 | +4 days | Objection or proof |
| 6 | +7 days | Last call, then stop |
The reasoning behind each:
Message 1 goes immediately — seconds, not minutes. They just interacted with your bot; attention is at its maximum and will never be this high again. Anything you promised on the button ships now.
Message 2 within hours, not a day. This is the biggest departure from email practice. A same-session follow-up while they still remember tapping is far more welcome than one arriving tomorrow from a bot they have forgotten.
Then decay. Gaps widen as the sequence progresses. Early messages have earned attention; later ones are spending it.
Six messages is usually the ceiling. Past that, block rate tends to outrun incremental conversion — you are trading durable reach for diminishing returns.
Determining your own intervals
Do not take the table above as measured truth. It is a reasoned starting point. Here is how to find your actual numbers.
Test one gap at a time. Split new subscribers, change a single interval, hold everything else. Sequences have too many variables to test more than one.
Judge on three metrics together:
- Click rate on the message
- Conversion by end of sequence
- Block rate
That third one is what makes this different from email testing. A tighter interval that lifts conversion 10% and doubles blocks is a bad trade you will not notice for months, because the cost is future reach rather than present revenue.
Watch time-to-click. Most Telegram clicks arrive quickly. If a message's clicks are still trickling in after six hours, your audience is not checking as fast as you assumed and your intervals can widen.
Respect the per-chat limit. Telegram advises no more than one message per second to a single chat. Multi-part messages need at least a second between them — and honestly, if you are sending three messages a second apart, consider sending one.
Timezones
The user object may carry language_code — the Bot API marks it Optional — which hints at region but is not a timezone, and you have to handle it being absent. You cannot read a user's timezone directly.
Two workable approaches:
Relative timing. Send at +4 hours from their signup, not at 9am absolute. Someone who joined at 11pm gets message two at 3am — which is why you clamp: no sends between, say, 10pm and 8am in your best-guess timezone, deferring to the morning.
Ask. For high-value sequences, a single button — "When should we message you? ☀️ Morning / 🌙 Evening" — is a small ask that solves it properly and doubles as an engagement signal in message one.
Things that reliably go wrong
Copying the email cadence exactly. Day 1 / day 3 / day 7 is too slow at the start. The first hours are where the attention is.
Sending message two the next day. By far the most common error. The window where they remember you is measured in hours.
No end. Sequences that never stop generate blocks indefinitely. End them, then move people to broadcasts.
Batching a sequence. Sending "day 3" to everyone on Tuesday turns a sequence into a broadcast and gives up the rate-limit advantage sequences have. Keep each subscriber on their own clock.
Ignoring block rate. It is the only metric that measures damage you cannot undo — a blocked user comes back only if they choose to unblock you, which is not something you can influence.
The principle
Faster at the start, slower later, and stop.
The first hour is worth more than the first week. The seventh message is almost never worth what it costs. And every interval decision is really a question about how much durable reach you are willing to spend for one more conversion.
Rate limits quoted from Telegram's Bot FAQ. Intervals above are reasoned from platform mechanics, not measured data — no Telegram-specific published timing benchmarks exist. Test against your own audience.